加入新推出的
Discord 社区,展开实时讨论,获得同行支持,并直接与 Meridian 团队互动!
设置 max_lag 形参
使用集合让一切井井有条
根据您的偏好保存内容并对其进行分类。
Meridian 模型允许时间点 \(t\) 的媒体影响时间点 \(t, t + 1, \dots , t + L\) 的 KPI,其中整数 \(L\) 是用户使用 ModelSpec
中的 max_lag
设置的超形参。媒体的影响可能会持续较长时间,超出 max_lag
所设置的时长。不过,由于几何衰减这一模型假设,媒体的滞后效应会趋近于零。
在实际应用中,max_lag
用于截断媒体产生影响的持续时间,因为这可以发挥积极的作用,包括提高模型收敛性、确保合理的模型运行时,以及最大限度地提高数据使用率(减少方差)。将 max_lag
的值保持在 2-10 的范围内,可以在优缺点之间达到良好的平衡。
提高 max_lag
的值并不一定意味着投资回报率估计值也会增加。原因之一是,如果时间点 \(t\)的媒体可以影响时间点 \(t+L\)的 KPI,这可能会削弱时间点 \(t+1, \dots , t+L\) 的媒体对时间点 \(t+L\)的 KPI 的影响。
如未另行说明,那么本页面中的内容已根据知识共享署名 4.0 许可获得了许可,并且代码示例已根据 Apache 2.0 许可获得了许可。有关详情,请参阅 Google 开发者网站政策。Java 是 Oracle 和/或其关联公司的注册商标。
最后更新时间 (UTC):2025-08-04。
[null,null,["最后更新时间 (UTC):2025-08-04。"],[[["\u003cp\u003eThe Meridian model assumes media impact on KPIs can extend over a period, defined by \u003ccode\u003emax_lag\u003c/code\u003e, with the effect diminishing over time due to geometric decay.\u003c/p\u003e\n"],["\u003cp\u003eWhile media impact can theoretically last longer than \u003ccode\u003emax_lag\u003c/code\u003e, it's truncated for practical reasons like model convergence, runtime, and data utilization.\u003c/p\u003e\n"],["\u003cp\u003eA \u003ccode\u003emax_lag\u003c/code\u003e value between 2 and 10 generally provides an optimal balance between model performance and efficiency.\u003c/p\u003e\n"],["\u003cp\u003eIncreasing \u003ccode\u003emax_lag\u003c/code\u003e might not result in higher ROI estimates, as it can redistribute the attributed impact across different media exposures over time.\u003c/p\u003e\n"]]],[],null,["# Set the max_lag parameter\n\nThe Meridian model allows for media at time \\\\(t\\\\) to affect the KPI at\ntimes \\\\(t, t + 1, \\\\dots , t + L\\\\) where the integer \\\\(L\\\\) is a\nhyperparameter set by the user using `max_lag` of `ModelSpec`. Media can\npotentially have a long effect that can go beyond `max_lag`. However, the lagged\neffect of media converges towards zero, due to the model assumption of geometric\ndecay.\n\nIn practice, `max_lag` is used to truncate how long media can have\nan effect because it has positive benefits including improved model\nconvergence, reasonable model runtimes, and maximizing data usage (reducing\nvariance). Keeping the `max_lag` in the 2-10 range leads to a good balance of\nthese advantages and disadvantages.\n\nIncreasing `max_lag` doesn't necessarily mean that ROI estimates\nwill also increase. One reason for this is because if the media at time \\\\(t\\\\)\ncan affect the KPI at time \\\\(t+L\\\\), this can take away from the effect of\nmedia at times \\\\(t+1, \\\\dots , t+L\\\\) on the KPI at time \\\\(t+L\\\\)."]]